Service function chaining is a mechanism for overriding the basic destination based forwarding that is typical of IP networks.
Hypervisor-based taps and network underlay-based packet mirroring are inefficient because they are configured outside of the context of normal Software-Designed Networking (“SDN”) and service chain management.
Thus, they are not available to the Virtual Machine (“VM”) layers to report metrics that would indicate when capacity levels are reached and/or whether additional resources are needed.
One other existing solution may be to place the vNIDS in-line in the service chain. While this allows for the requisite visibility regarding packet reordering, etc., which is lacking in hypervisor-based taps and network underlay-based packet mirroring, the addition of multiple monitoring elements (common in many network scenarios) introduces unnecessary additional latency in the service chain path.